Fedspeak is a personal project to quantify and analyze Federal Reserve Beige Book material. The goal is to use text mining and NLP techniques and think of qualitative information as a segway into quantitative indexes for regional economics and/or forecasting.
The Beige Book is a report published by the Federal Reserve 8 times a year and describes economic conditions across the United States for each of the 12 regional banks.
This respository contains:
- Documentation for how the code works and analysis of the project results. There's also a to-do here.
- Python code for text mining the Beige Book from the Minneapolis Federal Reserve website
- R code for finding sentiment of Beige Book reports by bank and as a time series to compare to regional economic indicators
